Traditionally, a scene, such as a movie or a sporting event, is captured from a single perspective in two dimensions. Accordingly, a rendering of the scene from second perspective, different from the original perspective, in two or three dimensions is not available to a user. As an example, at a given time, a scene in a movie is captured from a single perspective using a single camera and the perspective may be modified by either moving the camera or by switching to a different camera that is located in a different position than the original camera. The resulting scene is one that is captured from a single perspective at any given time and, thus, cannot be viewed or analyzed from a different perspective. Additionally, a user viewing a video feed that is captured using traditional techniques is limited to the perspectives that the video feed is captured in such that the perspective may not be changed based on user input or as a result of a predetermined setting. For example, a user may receive a video feed of a basketball game captured from the south end of the basketball court and may not have an option to switch the video feed to display the basketball game from the north end of the basketball court.